Intelligent adjusting feeding system and method for boiler blending combustion of low-carbon biomass fuel based on deep peak regulation

Abstract

The invention relates to the technical field of biomass fire coal blending combustion, in particular to an intelligent adjusting feeding system and method for boiler blending combustion of low-carbon biomass fuel based on deep peak shaving, and the system comprises a powder preparation storage unit, a transportation feeding unit, a collection unit, a prediction unit, a controller and a regulation and control unit. Wherein the prediction unit is used for cooperatively predicting the optimal pulverized coal feeding amount and the optimal biomass powder feeding amount based on a neural network prediction model and an optimization algorithm; and the regulation and control unit automatically regulates the opening degree of the valve according to the calculation result of the PID algorithm, so that the intelligent regulation of the feeding amount of the pulverized coal and the biomass powder is realized. According to the method, machine learning and an optimization algorithm are applied, accurate adjustment of the feeding amount of the pulverized coal and the biomass powder is achieved, the combustion efficiency is improved, and pollutant emission is reduced; in addition, the biomass blending combustion proportion is accurately adjusted through real-time monitoring and intelligent control technologies, the stable combustion capacity of the unit is remarkably improved under the low-load condition, and the peak regulation depth is remarkably optimized and improved.

Description

Technical Field

[0001] The present invention relates to the technical field of biomass-coal blending, and in particular to an intelligent regulating feed system and method for blending low-carbon biomass fuel in a boiler based on deep peak regulation. Background Art

[0002] In recent years, environmental pollution and global warming caused by excessive use of fossil fuels have attracted great attention from the international community. my country has proposed the goal of "carbon peak and carbon neutrality", which has put forward higher requirements for my country's energy structure and energy utilization. At present, a large amount of electricity in my country comes from coal-fired power generation. Since coal contains C, H, O, N, S and other elements, a large amount of carbon dioxide CO2 and nitrogen oxides NO will be produced during combustion. x and smoke pollutants, etc.

[0003] As the only renewable carbon energy at this stage, biomass fuel has the characteristics of low ash point, low sulfur point and high volatile matter. When biomass is mixed with coal, the combustion characteristics of the fuel can be optimized, making the fuel easier to ignite and burn stably. Especially when the boiler is running at low load, biomass blending can effectively improve the combustion stability of the boiler, reduce the occurrence of safety accidents such as flameout and deflagration, and further enhance the deep peak-shaving capacity of the boiler. At the same time, the growth process of biomass consumes carbon dioxide and realizes the natural cycle of carbon. The blending of biomass can effectively reduce the carbon emission intensity and pollutant emission level of coal-fired power units. In addition, by adjusting the blending ratio of biomass to coal, the output power of the boiler can be flexibly adjusted to adapt to the changes in the power grid load, providing a more reliable peak-shaving means for the boiler. It can be seen that biomass blending meets the strategic requirements of my country's energy transformation and sustainable development, and is one of the feasible and reliable technical routes for large-scale coal-fired power plants to achieve flexible adjustment of output power and promote the development of coal-fired power units towards green and low-carbon.

[0004] In addition, there are currently three main technical solutions for coupling coal-fired units with biomass power generation: direct combustion coupling, indirect coupling and parallel coupling. Direct combustion coupling has become the mainstream method adopted in China due to its relatively low transformation cost. However, direct combustion coupling technology directly feeds biomass fuel into a coal-fired boiler and mixes it with pulverized coal for combustion. There are some problems in this process. First, the traditional feeding system is mainly designed for coal and is difficult to adapt to the characteristics of biomass fuel. In actual applications, there are often phenomena such as uneven feeding of coal and biomass and incomplete combustion. Secondly, most of the current modified feeding systems lack the ability to control the ratio of coal and biomass feed in real time, ignoring the importance of the blending ratio to the combustion characteristics and stable operation of the boiler.

[0005] Therefore, there is an urgent need to develop an intelligent feeding system and method that can intelligently control coal and biomass feed and achieve effective mixing of the two. This is of vital importance to improving boiler combustion efficiency, reducing environmental pollution, enhancing unit peak-shaving flexibility, and improving unit peak-shaving depth. Summary of the invention

[0006] In view of the deficiencies in the above-mentioned prior art, the present invention provides an intelligent regulating feeding system and method for a boiler blended with low-carbon biomass fuel based on deep peak regulation, aiming to solve the problem that the traditional feeding system ignores the characteristics of biomass fuel and lacks real-time control of the blending ratio of coal and biomass fuel, thereby improving the combustion efficiency, operation stability and energy utilization efficiency of the boiler, reducing pollutant emissions, and achieving cleaner and more efficient energy utilization.

[0007] The present invention realizes precise adjustment of the feed amounts of coal powder and biomass powder by real-time monitoring of the operating status data of the boiler and by using machine learning and optimization algorithms, which not only significantly reduces the fluctuation caused by fuel feed and improves combustion efficiency, but also greatly reduces pollutant emissions and ensures stable operation of the boiler; in addition, through real-time monitoring and intelligent control technology, the biomass blending ratio is precisely adjusted, thereby significantly improving the stable combustion capacity of the unit under low load conditions, which not only enhances the peak-shaving flexibility of the unit, but also further taps the peak-shaving potential of the unit on the basis of ensuring stable operation, thereby achieving significant optimization and improvement of the peak-shaving depth.

[0008] The present invention provides an intelligent feed adjustment system for a boiler blended with low-carbon biomass fuel based on deep peak regulation, comprising:

[0009] A pulverizing and storage unit, used to pulverize coal and biomass powder and store them separately;

[0010] The transport and feeding unit is used to mix the pulverized coal and biomass powder evenly and then transport them to the boiler for combustion;

[0011] Multiple collection units collect boiler operation status data in real time at a sampling frequency of N minutes / times, where the boiler operation status data includes coal powder feed amount data, biomass powder feed amount data, combustion temperature data, boiler internal pressure data, CO concentration data, NO x Concentration data, O2 concentration data;

[0012] A data processing unit, used for preprocessing the data collected by each collection unit;

[0013] The prediction unit predicts NO based on the neural network prediction model and optimization algorithm. x The optimal coal powder feed rate and the optimal biomass powder feed rate when the concentration is the smallest;

[0014] The controller is used for the user to set initial parameters and transmit them to the control unit, and at the same time, based on the data uploaded by the second acquisition unit, it determines whether the combustion state in the boiler has reached a stable stage, and sends corresponding control instructions to the acquisition unit when the combustion state in the boiler has reached a stable stage, wherein the initial parameters include an initial coal powder feed amount and an initial biomass powder feed amount;

[0015] The control unit compares the data / initial parameters transmitted by the prediction unit with the data transmitted by the first acquisition unit after data preprocessing, and performs PID algorithm calculation, and automatically adjusts the valve opening according to the calculation result of the PID algorithm, thereby realizing intelligent adjustment of the feed amount of coal powder and biomass powder.

[0016] Preferably, the pulverizing storage unit comprises a coal powder storage subunit and a biomass powder storage subunit, wherein:

[0017] The pulverized coal storage subunit includes: a coal mill for grinding coal into pulverized coal; a pulverized coal bin for storing pulverized coal;

[0018] The biomass powder storage subunit includes: a pulverizer for grinding biomass particles into biomass powder; a biomass powder bin for storing biomass powder;

[0019] The transport and feeding unit includes a coal powder transport pipeline, a biomass powder transport pipeline and a vibration burner, wherein one end of the coal powder transport pipeline is connected to the coal powder bin, and the other end is connected to the vibration burner; one end of the biomass powder transport pipeline is connected to the biomass powder bin, and the other end is connected to the vibration burner; the vibration burner is arranged in the furnace of the boiler, and the biomass powder and coal powder are evenly mixed in the vibration burner and then sent into the boiler furnace for combustion.

[0020] Preferably, the biomass powder storage subunit also includes a cyclone separator, one end of which is connected to the pulverizer and the other end of which is connected to the biomass powder bin, for gas-solid separation of the biomass powder; the transport and feeding unit also includes a dryer, which is arranged in the biomass powder transport pipeline and is used to spray high-temperature air so that the high-moisture biomass absorbs heat during the high-temperature air mixing and high-speed circulation process, thereby drying the biomass powder.

[0021] Preferably, the plurality of acquisition units include:

[0022] The first collection unit is used to collect coal powder feed amount data and biomass powder feed amount data in real time;

[0023] The second acquisition unit is used to collect combustion temperature data and boiler internal pressure data in real time;

[0024] The third collection unit is used to collect CO and NO in flue gas in real time x , O2 concentration data;

[0025] The fourth collection unit is used to collect NO in flue gas in real time x Concentration data.

[0026] Preferably, the first acquisition unit is a mass flow sensor, specifically comprising: a first mass flow sensor, arranged on the coal powder conveying pipeline, for collecting coal powder feed amount data in real time; a second mass flow sensor, arranged on the biomass powder conveying pipeline, for collecting volume data and biomass powder feed amount data in real time;

[0027] The second acquisition unit specifically includes: a temperature sensor, which is arranged above the combustion chamber, for obtaining combustion temperature data; a pressure sensor, which is arranged at the boiler pressure gauge interface, for obtaining boiler internal pressure data;

[0028] The third and fourth collection units are both flue gas analyzers, which are installed at the collection pipe connected to the boiler exhaust pipe.

[0029] Preferably, the specific prediction steps of the prediction unit are:

[0030] S11, input the processed data of the second and third acquisition units into NO x The relationship between concentration and biomass powder feed amount, coal powder feed amount, boiler internal pressure, CO concentration, and O2 concentration t In the NO x The relationship between concentration and biomass powder feed amount and coal powder feed amount y′ t , where NO x The relationship between concentration and biomass powder feed amount, coal powder feed amount, boiler internal pressure, CO concentration, and O2 concentration t is used to predict the next moment NO x The concentration is obtained from the LSTM neural network prediction model, which is used to predict the next moment NO x The specific steps for constructing the LSTM neural network prediction model for concentration are as follows:

[0031] S111, obtaining historical operating status data of the boiler;

[0032] S112, preprocessing the historical operating status data of the boiler;

[0033] S113, dividing the boiler historical operation status data preprocessed in step S112 into a training set and a test set according to a ratio;

[0034] S114, construct a parameter matrix X based on the training set data, and perform standardization calculation on the parameter matrix to obtain a parameter standardization matrix X * ;

[0035] S115, standardize the parameter matrix X *Input into the LSTM neural network prediction model for training to obtain the initial prediction model;

[0036] S116: input the test set data into the initial prediction model for evaluation, and adjust the model parameters according to the evaluation results to obtain the parameters for predicting the next NO x The LSTM neural network prediction model of concentration, in which the evaluation method uses the absolute mean error MAE to evaluate the prediction results of each cross-validation test data;

[0037] S117, from the prediction of the next moment NO x NO was extracted from the LSTM neural network prediction model of concentration x The relationship between concentration and biomass powder feed amount, coal powder feed amount, boiler internal pressure, CO concentration, and O2 concentration t , the relation y t As shown below:

[0038] y t =W y ×h t +b y

[0039] In the formula, y t Indicates the next moment NO predicted at time step t x Concentration value, W y is the weight matrix of the linear layer, b y is the bias term of the linear layer; h t represents the hidden state at time step t, which is used to capture the input information up to the current time step, that is, the historical operating status data of the boiler;

[0040] S12, y′ t As the particle fitness value calculation formula, the particle swarm optimization algorithm PSO is used to find NO x The optimal coal powder feed rate and the optimal biomass powder feed rate at the minimum concentration are as follows:

[0041] S121. Define the particle as x i =[biomass powder feed amount, coal powder feed amount], and set the number of iterations;

[0042] S122, initializing the particle swarm, and at the same time initially setting the random position and speed of the particle swarm;

[0043] S123. Use formula y′ t Calculate the fitness value of each particle;

[0044] S124, for each particle, compare its fitness value with the fitness value of the best position it has experienced, if it is better, take it as the current individual optimal position;

[0045] S125, for each particle, compare its fitness value with the fitness value of the best position experienced globally, if it is better, take it as the current global optimal position;

[0046] S126. Optimize the speed and position of the particles to generate new particles, where the speed and position update formulas of the particles are as follows:

[0047]

[0048]

[0049] Where ω is the inertia weight, represents the velocity of the ith particle in the tth generation, represents the velocity of the ith particle in the t+1th generation, c1 and c2 are acceleration constants, r1 and r2 are random coefficients between [0, 1], Gbest represents the optimal solution searched by the i-th particle up to the t-th generation. t It represents the optimal solution searched by the entire particle swarm up to generation t. represents the position of the ith particle in the tth generation, represents the position of the i-th particle in the t+1 generation;

[0050] S127. Determine whether the maximum number of iterations has been reached. If so, stop updating and output the global optimal solution; otherwise, continue to execute steps S123-S127.

[0051] Preferably, the data processing unit and the preprocessing in step S112 specifically include: outlier detection, outlier processing, filtering and normalization, wherein the outlier processing includes outlier removal and outlier filling;

[0052] The data collected by the fourth collection unit is pre-processed by the data processing unit and then transmitted to the controller for storage; similarly, the prediction unit also x The optimal coal powder feed amount and the optimal biomass powder feed amount when the concentration is the smallest are converted into a coal powder to biomass powder feed ratio and then transmitted to the controller for storage.

[0053] Preferably, when the combustion state in the boiler reaches a stable stage, the controller sends an instruction to the second and third acquisition units to transmit data to the data processing unit, wherein the specific process of the controller judging whether the combustion state in the boiler has reached a stable stage based on the data uploaded by the second acquisition unit is: the data uploaded by the second acquisition unit is plotted into a curve graph, and the overall fluctuation amplitude of the data points is observed. When the fluctuation amplitude of the data points gradually decreases until the fluctuation tends to be flat, it is considered that the combustion state in the furnace has reached a stable stage;

[0054] In addition, when the biomass powder bin and coal powder bin are insufficient, the controller is also used to send an operation instruction to the powder making and storage unit to drive the coal mill and crusher to start grinding and storing coal powder and biomass powder.

[0055] Preferably, the regulating unit is an intelligent control valve, wherein the intelligent control valve includes a coal powder bin intelligent control valve and a biomass powder bin intelligent control valve, the coal powder bin intelligent control valve is arranged at the connection between the powder transport pipeline and the coal powder bin, and the biomass powder bin intelligent control valve is arranged at the connection between the biomass powder transport pipeline and the biomass powder bin;

[0056] In addition, the specific control process of the control unit is:

[0057] When the combustion state in the boiler has not reached a stable stage, the pulverized coal bin intelligent control valve compares the initial pulverized coal feed amount with the pulverized coal feed amount transmitted by the first acquisition unit after data preprocessing, and performs PID algorithm calculation, and automatically adjusts the valve opening according to the calculation results of the PID algorithm; at the same time, the biomass powder bin intelligent control valve compares the initial biomass powder feed amount with the biomass powder feed amount transmitted by the first acquisition unit, and performs PID algorithm calculation, and automatically adjusts the valve opening according to the calculation results of the PID algorithm, and through the synchronous adjustment of the pulverized coal bin intelligent control valve and the biomass powder bin intelligent control valve, the intelligent adjustment of the pulverized coal and biomass powder feed amounts is realized when the combustion state in the boiler has not reached a stable stage;

[0058] When the combustion state in the boiler has not reached a stable stage, the pulverized coal bin intelligent control valve compares the optimal pulverized coal feed amount transmitted by the prediction unit with the pulverized coal feed amount transmitted by the first acquisition unit, and performs PID algorithm calculation, and automatically adjusts the valve opening according to the calculation results of the PID algorithm; at the same time, the biomass powder bin intelligent control valve compares the optimal biomass powder feed amount transmitted by the prediction unit with the biomass powder feed amount transmitted by the first acquisition unit, and performs PID algorithm calculation, and automatically adjusts the valve opening according to the calculation results of the PID algorithm, through the synchronous adjustment of the pulverized coal bin intelligent control valve and the biomass powder bin intelligent control valve, thereby realizing intelligent adjustment of the pulverized coal and biomass powder feed amounts when the combustion state in the boiler reaches a stable stage.

[0059] The present invention also provides an intelligent feed adjustment method for a boiler blended with low-carbon biomass fuel based on deep peak shaving, which is applied to an intelligent feed adjustment system for a boiler blended with low-carbon biomass fuel based on deep peak shaving, and is characterized in that the specific steps are:

[0060] 1) When the combustion state in the boiler has not reached the stable stage, the specific steps are:

[0061] S21, the staff sets the initial parameters in the controller according to actual needs and transmits them to the control unit;

[0062] S22, the first and second acquisition units transmit the data to the data processing unit for preprocessing, and then the first acquisition unit transmits the data to the control unit, and the second acquisition unit transmits the data to the controller;

[0063] S23, the control unit compares the initial parameters with the data transmitted by the first acquisition unit, and performs PID algorithm calculation, automatically adjusts the valve opening according to the calculation result of the PID algorithm, and then transports the evenly mixed coal powder and biomass powder to the boiler through the transport feeding unit for combustion; at the same time, the controller determines whether the combustion state in the boiler has reached a stable stage based on the data uploaded by the second acquisition unit;

[0064] S24, repeat steps S22-S23;

[0065] S25, when the combustion state in the boiler reaches a stable stage, the controller 14 sends instructions to the second and third acquisition units to transmit data to the data processing unit, and stops executing step S24;

[0066] 2) When the combustion state in the boiler reaches a stable stage, the specific steps are:

[0067] S31, the first, second and third acquisition units transmit data to the data processing unit for preprocessing, then the first acquisition unit transmits the data to the control unit, and the second and third acquisition units transmit the data to the prediction unit;

[0068] S32, the prediction unit predicts NO based on the neural network prediction model and optimization algorithm x The optimal coal powder feed amount and the optimal biomass powder feed amount with the minimum concentration are then transmitted to the control unit;

[0069] S33, the control unit compares the data transmitted by the prediction unit with the data transmitted by the first acquisition unit, and performs PID algorithm calculation, automatically adjusts the valve opening according to the calculation result of the PID algorithm, and then transports the evenly mixed coal powder and biomass powder to the boiler through the transport feeding unit for combustion;

[0070] S34, repeating steps S31-S33 to complete the intelligent adjustment of the feed amount of pulverized coal and biomass powder when the combustion state in the boiler reaches a stable stage;

[0071] S35. After the boiler stops burning, stop executing step S34.

[0072] Compared with the prior art, the present invention has the following beneficial effects:

[0073] 1. The present invention not only solves the problem of traditional feeding system ignoring the characteristics of biomass fuel by grinding and storing coal powder and biomass powder separately, but also avoids the accumulation and floating of biomass particles in the coal mill, reduces the interference of biomass powder on the coal powder conveying system, reduces the risk of blockage, and ensures the stable operation of the system; in addition, it is convenient to flexibly adjust the ratio of coal powder and biomass powder according to the combustion requirements of the boiler and the characteristics of coal quality, so that they can be mixed and burned more fully during combustion to achieve the best combustion effect, and at the same time can reduce the content of unburned carbon, thereby reducing the emission of pollutants.

[0074] 2. The present invention uses high-temperature air generated by a dryer to dry the biomass powder, which not only has a good drying effect and is conducive to the full combustion of the biomass powder, but also solves the problem of the traditional feeding system ignoring the characteristics of biomass fuel; in addition, the air sprayed by the dryer has a higher flow rate, which is conducive to the transmission of biomass powder.

[0075] 3. The present invention utilizes the vibration principle and uses a vibrating burner to fully mix the coal powder and the biomass powder inside the burner, which significantly improves the mixing effect of the coal powder and the biomass powder, improves the combustion efficiency of the fuel, and reduces the incomplete combustion phenomenon and the emission of pollutants. In addition, the vibration effect of the vibrating burner is used to effectively separate the slag from the fuel, avoiding uneven fuel stacking and mixing of the fuel and slag, and further improving the combustion efficiency of the fuel.

[0076] 4. The present invention realizes precise adjustment of the feed amount of coal powder and biomass powder by real-time monitoring of the operating status data of the boiler and using machine learning and optimization algorithms. It not only significantly reduces the fluctuation caused by fuel feed and improves combustion efficiency, but also greatly reduces pollutant emissions and ensures the stable operation of the boiler. It provides strong support for achieving cleaner and more efficient energy utilization and has broad application prospects.

[0077] 5. The present invention uses real-time monitoring and intelligent control technology to accurately adjust the biomass blending ratio, thereby significantly improving the stable combustion capacity of the unit under low-load conditions. It not only enhances the peak-shaving flexibility of the unit, but also further taps the peak-shaving potential of the unit on the basis of ensuring stable operation, and achieves significant optimization and improvement of the peak-shaving depth. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0079] Figure 1 A schematic block diagram of an intelligent feed regulation system for a boiler blended with low-carbon biomass fuel based on deep peak regulation provided by the present invention;

[0080] Figure 2 A working principle diagram of the intelligent feed adjustment system for the boiler blended with low-carbon biomass fuel based on deep peak regulation provided by the present invention. DETAILED DESCRIPTION

[0081] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0082] In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific position, be constructed and operated in a specific position, and therefore cannot be understood as limiting the present invention; the terms "first", "second", and "third" are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance. In addition, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be an indirect connection through an intermediate medium, or it can be a connection between the two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0083] like Figure 1-2 As shown, the present invention provides an intelligent regulating feeding system for a boiler blended with low-carbon biomass fuel based on deep peak regulation, the system comprising a pulverizing storage unit, a transport feeding unit, multiple collection units, a data processing unit, a prediction unit, a controller 14 and a control unit.

[0084] The pulverizing storage unit of the present invention comprises a coal powder storage subunit and a biomass powder storage subunit, which are used for respectively preparing coal powder and biomass powder and storing them separately.

[0085] In the present application, the pulverized coal storage subunit includes: a coal mill 1, which is used to grind coal into pulverized coal; and a pulverized coal bin 3, which is used to store pulverized coal.

[0086] In the present application, the biomass powder storage subunit includes: a grinder 2 for grinding biomass particles into biomass powder; and a biomass powder bin 4 for storing biomass powder.

[0087] Since coal powder and biomass powder have different physical and chemical properties, mixed storage may lead to quality degradation, such as the ignition performance and burnout performance of coal powder may be affected, and biomass powder may cause arching, bridging, and compaction due to its strong hygroscopicity. Separate storage can avoid these unfavorable factors and ensure the quality of the powder. In addition, the bulk density of biomass powder is low and its fluidity is poor, which is easy to cause blockage during storage and transportation. Therefore, in this application, biomass powder and coal powder are ground and stored separately, which not only avoids the accumulation and floating of biomass particles in the coal mill, reduces the interference of biomass powder on the coal powder transportation system, reduces the risk of blockage, ensures the stable operation of the system, but also facilitates the flexible adjustment of the ratio of coal powder and biomass powder according to the combustion requirements of the boiler and the coal quality characteristics, so that they can be more fully mixed and burned during combustion to achieve the best combustion effect, while also reducing the content of unburned carbon, thereby reducing pollutant emissions.

[0088] Preferably, the biomass powder storage subunit further includes a cyclone separator 5, one end of which is connected to the pulverizer 2, and the other end of which is connected to the biomass powder bin 4, for performing gas-solid separation on the biomass powder.

[0089] The transport and feeding unit involved in the present invention is used for evenly mixing coal powder and biomass powder and then transporting them to a boiler for combustion.

[0090] In the present application, the transport and feeding unit includes a coal powder transport pipeline 9, a biomass powder transport pipeline 8 and a vibration burner 13, wherein one end of the coal powder transport pipeline 9 is connected to the coal powder bin 3, and the other end is connected to the vibration burner 13; one end of the biomass powder transport pipeline 8 is connected to the biomass powder bin 4, and the other end is connected to the vibration burner 13; the vibration burner 13 is arranged in the furnace of the boiler, and the biomass powder and coal powder are mixed evenly in the vibration burner 13 and then sent into the boiler furnace for combustion.

[0091] The vibration principle is utilized in the present application, and the coal powder and biomass powder are fully mixed inside the burner through a vibrating burner, which significantly improves the mixing effect of the coal powder and the biomass powder, improves the combustion efficiency of the fuel, and reduces incomplete combustion and pollutant emissions. In addition, the slag and the fuel are effectively separated by the vibration effect of the vibrating burner, avoiding uneven fuel stacking and mixing of the fuel and slag, and further improving the combustion efficiency of the fuel.

[0092] Preferably, the transport and feeding unit further comprises a dryer 12, which is arranged in the biomass powder transport pipeline 8 and is used for spraying high-temperature air so that the high-moisture biomass absorbs heat during the high-temperature air mixing and high-speed circulation process, thereby drying the biomass powder.

[0093] Since wet biomass contains more water and has a lower density, it is easy to clog pipes and equipment during transportation, and it is easy to cause unstable combustion due to water evaporation during combustion. Therefore, in this application, a dryer 12 is set in the biomass powder transportation pipe 8 to dry the biomass powder, so that the dried biomass powder can flow more smoothly, avoid pipe clogging, and make the combustion process more efficient and stable.

[0094] The multiple acquisition units involved in the present invention collect boiler operation status data in real time at a sampling frequency of N minutes / times, wherein the boiler operation status data includes coal powder feed amount data, biomass powder feed amount data, combustion temperature data, boiler internal pressure data, CO concentration data, NO x Concentration data, O2 concentration data.

[0095] In this application, multiple acquisition units include:

[0096] The first collection unit is used to collect coal powder feed amount data and biomass powder feed amount data in real time;

[0097] The second acquisition unit is used to collect combustion temperature data and boiler internal pressure data in real time;

[0098] The third collection unit is used to collect CO and O2 concentration data in the flue gas in real time;

[0099] The fourth collection unit is used to collect NO in flue gas in real time x Concentration data.

[0100] It should be noted that in the embodiments of the present application, the feed amount refers to the feed amount of coal powder / biomass powder per unit time, rather than the total amount fed.

[0101] In the embodiment of the present application, the first acquisition unit is a mass flow sensor, specifically comprising:

[0102] A first mass flow sensor 11 is provided on the pulverized coal conveying pipeline 9 and is used to collect pulverized coal feed amount data in real time;

[0103] The second mass flow sensor 10 is arranged on the biomass powder conveying pipeline 8 and is used for collecting the quantity data and the biomass powder feed quantity data in real time.

[0104] In the embodiment of the present application, the second acquisition unit specifically includes:

[0105] A temperature sensor 15 is disposed above the combustion chamber and is used to obtain combustion temperature data;

[0106] The pressure sensor 16 is arranged at the boiler pressure gauge interface and is used to obtain the internal pressure data of the boiler.

[0107] In the embodiment of the present application, the third and fourth collection units are both flue gas analyzers 17, which are arranged at a collection pipe connected to the boiler exhaust pipe.

[0108] It should be noted that the data collected by the fourth acquisition unit is preprocessed by the data processing unit and then transmitted to the controller 14 for storage. In the embodiment of the present application, the data collected by the fourth acquisition unit is used for training the LSTM neural network prediction model, and does not participate in the intelligent adjustment of the feed amount of coal powder and biomass powder.

[0109] Preferably, the temperature sensor 15 is an infrared temperature sensor, the pressure sensor 16 is a diaphragm pressure sensor, and the flue gas analyzer 17 is an infrared flue gas analyzer.

[0110] The data processing unit involved in the present invention is used for preprocessing the data collected by each collection unit.

[0111] In the embodiment of the present application, the preprocessing in the data processing unit specifically includes: outlier detection, outlier processing, filtering processing and normalization processing, wherein the outlier processing includes outlier removal and outlier filling.

[0112] The prediction unit of the present invention predicts NO based on the neural network prediction model and the optimization algorithm. x The optimal coal powder feed rate and the optimal biomass powder feed rate when the concentration is minimum.

[0113] In this application, the specific prediction steps of the prediction unit are:

[0114] S11, input the processed data of the second and third acquisition units into NO x The relationship between concentration and biomass powder feed amount, coal powder feed amount, boiler internal pressure, CO concentration, and O2 concentration t In the NO x The relationship between concentration and biomass powder feed amount and coal powder feed amount y′ t .

[0115] In the present application, the NO x The relationship between concentration and biomass powder feed amount, coal powder feed amount, boiler internal pressure, CO concentration, and O2 concentration t is used to predict the next moment NO x The concentration is obtained from the LSTM neural network prediction model.

[0116] It should be noted that the "next moment" mentioned in the present invention is not the well-known moment, ie, 15 minutes.

[0117] Preferably, the method for predicting the next moment NO x The specific steps for constructing the LSTM neural network prediction model for concentration are as follows:

[0118] S111. Obtain historical operating status data of the boiler.

[0119] In the embodiment of the present application, the historical operation status data of the boiler includes coal powder feed amount data, biomass powder feed amount data, combustion temperature data, boiler internal pressure data, CO concentration data, NO x Concentration data, O2 concentration data.

[0120] S112. Preprocess the boiler historical operating status data.

[0121] Preferably, the preprocessing in step S112 specifically includes: detection, outlier processing, filtering and normalization, wherein the outlier processing includes outlier removal and outlier filling.

[0122] S113, dividing the boiler historical operating status data preprocessed in step S112 into a training set and a test set according to a proportion.

[0123] It should be noted that, in this embodiment, in order to ensure the stability and reproducibility of the training model, a 10-fold cross validation was performed, and the training set and the test set accounted for 80% and 20% of the entire sample, respectively.

[0124] S114, construct a parameter matrix X based on the training set data, and perform standardization calculation on the parameter matrix to obtain a parameter standardization matrix X * .

[0125] S115, standardize the parameter matrix X * Input into the LSTM neural network prediction model for training to obtain the initial prediction model.

[0126] In this application, the training formula of the LSTM neural network prediction model is as follows:

[0127]

[0128] In the formula, f t is the output of the forget gate, σ is the sigmoid activation function, W f is the weight matrix of the forget gate, b f is the bias term of the forget gate, i t is the output of the input gate, W i is the weight matrix of the input gate, b iis the bias term of the input gate, is a candidate memory unit, tanh is the hyperbolic tangent activation function, W C is the weight matrix of the memory unit, b C is the bias term of the memory unit, C t-1 , C t is the unit state of the previous time step and the current time step, o t is the output gate activation value, W o is the weight matrix of the output gate, b o is the bias term of the output gate, x t represents the input feature vector; h t-1 、h t is the hidden state of the previous time step and the current time step, which is used for output.

[0129] In the present application embodiment, x t Including coal powder feed amount data, biomass powder feed amount data, combustion temperature data, boiler internal pressure data, CO concentration data, and O2 concentration data.

[0130] In the embodiment of the present application, the coal powder feed amount data, biomass powder feed amount data, combustion temperature data, boiler internal pressure data, CO concentration data, and O2 concentration data are used as input parameters of the LSTM neural network prediction model to predict the NO concentration at the next moment. x concentration.

[0131] S116: input the test set data into the initial prediction model for evaluation, and adjust the model parameters according to the evaluation results to obtain the parameters for predicting the next NO x The LSTM neural network prediction model of concentration, in which the evaluation method uses the absolute mean error MAE to evaluate the prediction results of each cross-validation test data. The absolute mean error MAE formula is as follows:

[0132]

[0133] In the formula, n represents the number of samples, y i represents the true value of the i-th sample, Represents the predicted value of the i-th sample.

[0134] S117, from the prediction of the next moment NO x NO was extracted from the LSTM neural network prediction model of concentration x The relationship between concentration and biomass powder feed amount, coal powder feed amount, boiler internal pressure, CO concentration, and O2 concentration t , where the relationship y t As shown below:

[0135] yt =W y ×h t +b y

[0136] In the formula, y t Indicates the next moment NO predicted at time step t x Concentration value, W y is the weight matrix of the linear layer, b y is the bias term of the linear layer; h t Represents the hidden state at time step t, which is used to capture the input information up to the current time step, that is, the historical operating status data of the boiler.

[0137] It should be noted that in the process of extracting the formula, h t Capture the boiler's historical operating status data and predict the next NO x concentration value, thereby obtaining NO x The relationship between concentration and biomass powder feed amount, coal powder feed amount, boiler internal pressure, CO concentration, and O2 concentration t However, when the system is actually running, the coal powder feed amount data and biomass powder feed amount data are set as unknown parameters, while the combustion temperature data, boiler internal pressure data, CO concentration data, NO x The concentration data and O2 concentration data are converted into NO x The relationship between concentration and biomass powder feed amount and coal powder feed amount y′ t .

[0138] S12, y′ t As the particle fitness value calculation formula, the particle swarm optimization algorithm PSO is used to find NO x The optimal coal powder feed rate and the optimal biomass powder feed rate when the concentration is minimum.

[0139] In the embodiment of the present application, the specific steps of step S12 are:

[0140] S121. Define the particle as x i =[biomass powder feed amount, coal powder feed amount], and set the number of iterations.

[0141] S122, initializing the particle swarm and initially setting the random position and speed of the particle swarm.

[0142] S123. Use formula y′ t Calculate the fitness value of each particle.

[0143] In the present application, since it is necessary to find NO xThe optimal coal powder feed rate and the optimal biomass powder feed rate when the concentration is the smallest, so the smaller the particle fitness value, the better.

[0144] S124. For each particle, compare its fitness value with the fitness value of the best position it has experienced. If it is better, use it as the current individual optimal position.

[0145] S125. For each particle, compare its fitness value with the fitness value of the best position experienced globally. If it is better, use it as the current global optimal position.

[0146] S126. Optimize the speed and position of the particles to generate new particles, where the speed and position update formulas of the particles are as follows:

[0147]

[0148]

[0149] Where ω is the inertia weight, represents the velocity of the ith particle in the tth generation, represents the velocity of the ith particle in the t+1th generation, c1 and c2 are acceleration constants, r1 and r2 are random coefficients between [0, 1], Gbest represents the optimal solution searched by the i-th particle up to the t-th generation. t It represents the optimal solution searched by the entire particle swarm up to generation t. represents the position of the ith particle in the tth generation, represents the position of the i-th particle in the t+1th generation.

[0150] S127. Determine whether the maximum number of iterations has been reached. If so, stop updating and output the global optimal solution; otherwise, continue to execute steps S123-S127.

[0151] Preferably, the prediction unit also sets NO x The optimal coal powder feed amount and the optimal biomass powder feed amount when the concentration is the smallest are converted into a coal powder to biomass powder feed ratio and then transmitted to the controller 14 for storage.

[0152] The controller 14 involved in the present invention is used for the user to set initial parameters and transmit them to the control unit. At the same time, based on the data uploaded by the second acquisition unit, it determines whether the combustion state in the boiler has reached a stable stage, and sends corresponding control instructions to the acquisition unit when the combustion state in the boiler reaches a stable stage, wherein the initial parameters include the initial coal powder feed amount and the initial biomass powder feed amount.

[0153] In the present application, when the combustion state in the boiler reaches a stable stage, the controller 14 sends instructions to the second and third acquisition units to transmit data to the data processing unit.

[0154] Preferably, the specific process by which the controller 14 determines whether the combustion state in the boiler has reached a stable stage based on the data uploaded by the second acquisition unit is: plotting the data uploaded by the second acquisition unit into a curve graph, observing the overall fluctuation amplitude of the data points, and when the fluctuation amplitude of the data points gradually decreases until the fluctuation tends to be gentle, it is considered that the combustion state in the furnace has reached a stable stage.

[0155] In addition, when the biomass powder bin 4 and the coal powder bin 3 are insufficient, the controller 14 is also used to send an operation instruction to the powder making and storage unit to drive the coal mill 1 and the crusher 2 to start grinding and storing coal powder and biomass powder.

[0156] It should be noted that for the observation of the powder storage amount in the biomass powder bin 4 and the coal powder bin 3, staff can be arranged to regularly patrol the biomass powder bin 4 and the coal powder bin 3 to check the powder storage situation in the powder bin. High-precision sensors can be installed in the biomass powder bin 4 and the coal powder bin 3 to collect the weight or height changes of the powder in the powder bin in real time, so as to accurately calculate the powder storage amount. The powder storage situation in the powder bin can also be remotely monitored through terminal devices such as mobile phones and computers. In this application, there is no specific restriction on the method of observing the powder storage amount in the biomass powder bin 4 and the coal powder bin 3.

[0157] The control unit involved in the present invention compares the data / initial parameters transmitted by the prediction unit with the data transmitted by the first acquisition unit after data preprocessing, and performs PID algorithm calculation, and automatically adjusts the valve opening according to the calculation result of the PID algorithm, thereby realizing intelligent adjustment of the feed amount of coal powder and biomass powder.

[0158] In the present application, the regulating unit is an intelligent control valve, wherein the intelligent control valve includes a coal powder bin intelligent control valve 6 and a biomass powder bin intelligent control valve 7. The coal powder bin intelligent control valve 6 is arranged at the connection point between the powder transport pipeline 9 and the coal powder bin 3, and the biomass powder bin intelligent control valve 7 is arranged at the connection point between the biomass powder transport pipeline 8 and the biomass powder bin 4.

[0159] In the embodiment of the present application, the specific control process of the control unit is:

[0160] When the combustion state in the boiler has not reached a stable stage, the pulverized coal bin intelligent control valve 6 compares the initial pulverized coal feed amount with the pulverized coal feed amount transmitted by the first acquisition unit after data preprocessing, and performs PID algorithm calculation, and automatically adjusts the valve opening according to the calculation result of the PID algorithm; at the same time, the biomass powder bin intelligent control valve 7 compares the initial biomass powder feed amount with the biomass powder feed amount transmitted by the first acquisition unit, and performs PID algorithm calculation, and automatically adjusts the valve opening according to the calculation result of the PID algorithm, through the synchronous adjustment of the pulverized coal bin intelligent control valve 6 and the biomass powder bin intelligent control valve 7, thereby realizing the intelligent adjustment of the pulverized coal and biomass powder feed amounts when the combustion state in the boiler has not reached a stable stage;

[0161] When the combustion state in the boiler has not reached a stable stage, the pulverized coal bin intelligent control valve 6 compares the optimal pulverized coal feed amount transmitted by the prediction unit with the pulverized coal feed amount transmitted by the first acquisition unit, and performs PID algorithm calculation, and automatically adjusts the valve opening according to the calculation results of the PID algorithm; at the same time, the biomass powder bin intelligent control valve 7 compares the optimal biomass powder feed amount transmitted by the prediction unit with the biomass powder feed amount transmitted by the first acquisition unit, and performs PID algorithm calculation, and automatically adjusts the valve opening according to the calculation results of the PID algorithm. Through the synchronous adjustment of the pulverized coal bin intelligent control valve 6 and the biomass powder bin intelligent control valve 7, intelligent adjustment of the pulverized coal and biomass powder feed amounts is achieved when the combustion state in the boiler reaches a stable stage.

[0162] The present invention also provides an intelligent feed adjustment method for a boiler blended with low-carbon biomass fuel based on deep peak shaving, which is applied to an intelligent feed adjustment system for a boiler blended with low-carbon biomass fuel based on deep peak shaving, and the specific steps are:

[0163] 1) When the combustion state in the boiler has not reached the stable stage, the specific steps are:

[0164] S21, the staff sets the initial parameters in the controller 14 according to actual needs and transmits them to the control unit;

[0165] S22, the first and second acquisition units transmit the data to the data processing unit for preprocessing, and then the first acquisition unit transmits the data to the control unit, and the second acquisition unit transmits the data to the controller 14;

[0166] S23, the control unit compares the initial parameters with the data transmitted by the first acquisition unit, and performs PID algorithm calculation, automatically adjusts the valve opening according to the calculation result of the PID algorithm, and then transports the evenly mixed coal powder and biomass powder to the boiler through the transport feeding unit for combustion; at the same time, the controller 14 determines whether the combustion state in the boiler has reached a stable stage based on the data uploaded by the second acquisition unit;

[0167] S24, repeat steps S22-S23;

[0168] S25. When the combustion state in the boiler reaches a stable stage, the controller 14 sends instructions to the second and third acquisition units to transmit data to the data processing unit, and stops executing step S24.

[0169] 2) When the combustion state in the boiler reaches a stable stage, the specific steps are:

[0170] S31, the first, second and third acquisition units transmit data to the data processing unit for preprocessing, then the first acquisition unit transmits the data to the control unit, and the second and third acquisition units transmit the data to the prediction unit;

[0171] S32, the prediction unit predicts NO based on the neural network prediction model and optimization algorithm x The optimal coal powder feed amount and the optimal biomass powder feed amount with the minimum concentration are then transmitted to the control unit;

[0172] S33, the control unit compares the data transmitted by the prediction unit with the data transmitted by the first acquisition unit, and performs PID algorithm calculation, automatically adjusts the valve opening according to the calculation result of the PID algorithm, and then transports the evenly mixed coal powder and biomass powder to the boiler through the transport feeding unit for combustion;

[0173] S34, repeating steps S31-S33 to complete the intelligent adjustment of the feed amount of pulverized coal and biomass powder when the combustion state in the boiler reaches a stable stage;

[0174] S35. After the boiler stops burning, stop executing step S34.

[0175] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An intelligent feed adjustment system for boilers burning low-carbon biomass fuels based on deep peak load regulation, characterized in that: include: A pulverizing and storage unit, used to pulverize coal and biomass powder and store them separately; The transport and feeding unit is used to mix the pulverized coal and biomass powder evenly and then transport them to the boiler for combustion; Multiple collection units collect boiler operation status data in real time at a sampling frequency of N minutes / times, where the boiler operation status data includes coal powder feed amount data, biomass powder feed amount data, combustion temperature data, boiler internal pressure data, CO concentration data, NO x Concentration data, O2 concentration data; A data processing unit, used for preprocessing the data collected by each collection unit; The prediction unit predicts NO based on the neural network prediction model and optimization algorithm. x The optimal coal powder feed rate and the optimal biomass powder feed rate when the concentration is the smallest; The controller is used for the user to set initial parameters and transmit them to the control unit, and at the same time, based on the data uploaded by the second acquisition unit, it determines whether the combustion state in the boiler has reached a stable stage, and sends corresponding control instructions to the acquisition unit when the combustion state in the boiler has reached a stable stage, wherein the initial parameters include an initial coal powder feed amount and an initial biomass powder feed amount; The control unit compares the data / initial parameters transmitted by the prediction unit with the data transmitted by the first acquisition unit after data preprocessing, and performs PID algorithm calculation, and automatically adjusts the valve opening according to the calculation result of the PID algorithm, thereby realizing intelligent adjustment of the feed amount of coal powder and biomass powder.

2. The intelligent feed adjustment system for boilers burning low-carbon biomass fuels based on deep peak load regulation according to claim 1 is characterized in that: The pulverizing storage unit includes a coal powder storage subunit and a biomass powder storage subunit, wherein: The pulverized coal storage subunit includes: a coal mill for grinding coal into pulverized coal; a pulverized coal bin for storing pulverized coal; The biomass powder storage subunit includes: a pulverizer for grinding biomass particles into biomass powder; a biomass powder bin for storing biomass powder; The transport and feeding unit includes a coal powder transport pipeline, a biomass powder transport pipeline and a vibration burner, wherein one end of the coal powder transport pipeline is connected to the coal powder bin, and the other end is connected to the vibration burner; one end of the biomass powder transport pipeline is connected to the biomass powder bin, and the other end is connected to the vibration burner; the vibration burner is arranged in the furnace of the boiler, and the biomass powder and coal powder are evenly mixed in the vibration burner and then sent into the boiler furnace for combustion.

3. The intelligent feed adjustment system for boilers burning low-carbon biomass fuels based on deep peak load regulation according to claim 2 is characterized in that: The biomass powder storage subunit also includes a cyclone separator, one end of which is connected to the pulverizer and the other end is connected to the biomass powder bin, for gas-solid separation of the biomass powder; the transport and feeding unit also includes a dryer, which is arranged in the biomass powder transport pipeline and is used to spray high-temperature air so that the high-moisture biomass absorbs heat during the high-temperature air mixing and high-speed circulation process, thereby drying the biomass powder.

4. The intelligent feed adjustment system for boilers burning low-carbon biomass fuels based on deep peak load regulation according to claim 3 is characterized in that: Multiple acquisition units include: The first collection unit is used to collect coal powder feed amount data and biomass powder feed amount data in real time; The second acquisition unit is used to collect combustion temperature data and boiler internal pressure data in real time; The third collection unit is used to collect CO and NO in flue gas in real time x , O2 concentration data; The fourth collection unit is used to collect NO in flue gas in real time x Concentration data.

5. The intelligent feed adjustment system for boilers burning low-carbon biomass fuels based on deep peak load regulation according to claim 4 is characterized in that: The first acquisition unit is a mass flow sensor, specifically comprising: a first mass flow sensor, arranged on the coal powder conveying pipeline, for collecting coal powder feed amount data in real time; a second mass flow sensor, arranged on the biomass powder conveying pipeline, for collecting volume data and biomass powder feed amount data in real time; The second acquisition unit specifically includes: a temperature sensor, which is arranged above the combustion chamber, for obtaining combustion temperature data; a pressure sensor, which is arranged at the boiler pressure gauge interface, for obtaining boiler internal pressure data; The third and fourth collection units are both flue gas analyzers, which are installed at the collection pipe connected to the boiler exhaust pipe.

6. The intelligent feed adjustment system for boilers burning low-carbon biomass fuels based on deep peak load regulation according to claim 5 is characterized in that: The specific prediction steps of the prediction unit are: S11, input the processed data of the second and third acquisition units into NO x The relationship between concentration and biomass powder feed amount, coal powder feed amount, boiler internal pressure, CO concentration, and O2 concentration t In the NO x The relationship between concentration and biomass powder feed amount and coal powder feed amount y′ t , where NO x The relationship between concentration and biomass powder feed amount, coal powder feed amount, boiler internal pressure, CO concentration, and O2 concentration t is used to predict the next moment NO x The concentration is obtained from the LSTM neural network prediction model, which is used to predict the next moment NO x The specific steps for constructing the LSTM neural network prediction model for concentration are as follows: S111, obtaining historical operating status data of the boiler; S112, preprocessing the historical operating status data of the boiler; S113, dividing the boiler historical operation status data preprocessed in step S112 into a training set and a test set according to a ratio; S114, construct a parameter matrix X based on the training set data, and perform standardization calculation on the parameter matrix to obtain a parameter standardization matrix X * ; S115, standardize the parameter matrix X * Input into the LSTM neural network prediction model for training to obtain the initial prediction model; S116: input the test set data into the initial prediction model for evaluation, and adjust the model parameters according to the evaluation results to obtain the parameters for predicting the next NO x The LSTM neural network prediction model of concentration, in which the evaluation method uses the absolute mean error MAE to evaluate the prediction results of each cross-validation test data; S117, from the prediction of the next moment NO x NO was extracted from the LSTM neural network prediction model of concentration x The relationship between concentration and biomass powder feed amount, coal powder feed amount, boiler internal pressure, CO concentration, and O2 concentration t , the relation y t As shown below: y t =W y ×h t +b y In the formula, y t Indicates the next moment NO predicted at time step t x Concentration value, W y is the weight matrix of the linear layer, b y is the bias term of the linear layer; h t represents the hidden state at time step t, which is used to capture the input information up to the current time step, that is, the historical operating status data of the boiler; S12, y′ t As the particle fitness value calculation formula, the particle swarm optimization algorithm PSO is used to find NO x The optimal coal powder feed rate and the optimal biomass powder feed rate at the minimum concentration are as follows: S121. Define the particle as x i =[biomass powder feed amount, coal powder feed amount], and set the number of iterations; S122, initializing the particle swarm, and at the same time initially setting the random position and speed of the particle swarm; S123. Use formula y′ t Calculate the fitness value of each particle; S124, for each particle, compare its fitness value with the fitness value of the best position it has experienced, if it is better, take it as the current individual optimal position; S125, for each particle, compare its fitness value with the fitness value of the best position experienced globally, if it is better, take it as the current global optimal position; S126. Optimize the speed and position of the particles to generate new particles, where the speed and position update formulas of the particles are as follows: Where ω is the inertia weight, represents the velocity of the ith particle in the tth generation, represents the velocity of the ith particle in the t+1th generation, c1 and c2 are acceleration constants, r1 and r2 are random coefficients between [0,1], Gbest represents the optimal solution searched by the i-th particle up to the t-th generation. t represents the optimal solution searched by the entire particle swarm up to generation t, represents the position of the ith particle in the tth generation, represents the position of the i-th particle in the t+1 generation; S127. Determine whether the maximum number of iterations has been reached. If so, stop updating and output the global optimal solution; otherwise, continue to execute steps S123-S127.

7. The intelligent feed adjustment system for boilers burning low-carbon biomass fuels based on deep peak load regulation according to claim 6 is characterized in that: The data processing unit and the preprocessing in step S112 specifically include: outlier detection, outlier processing, filtering and normalization, wherein the outlier processing includes outlier removal and outlier filling; The data collected by the fourth collection unit is pre-processed by the data processing unit and then transmitted to the controller for storage; similarly, the prediction unit also x The optimal coal powder feed amount and the optimal biomass powder feed amount when the concentration is the smallest are converted into a coal powder to biomass powder feed ratio and then transmitted to the controller for storage.

8. The intelligent feed adjustment system for boilers burning low-carbon biomass fuels based on deep peak load regulation according to claim 7 is characterized in that: When the combustion state in the boiler reaches a stable stage, the controller sends an instruction to the second and third acquisition units to transmit data to the data processing unit, wherein the specific process of the controller judging whether the combustion state in the boiler has reached a stable stage based on the data uploaded by the second acquisition unit is as follows: the data uploaded by the second acquisition unit is plotted into a curve graph, and the overall fluctuation amplitude of the data points is observed. When the fluctuation amplitude of the data points gradually decreases until the fluctuation tends to be flat, it is considered that the combustion state in the furnace has reached a stable stage; In addition, when the biomass powder bin and coal powder bin are insufficient, the controller is also used to send an operation instruction to the powder making and storage unit to drive the coal mill and crusher to start grinding and storing coal powder and biomass powder.

9. The intelligent feed adjustment system for boilers burning low-carbon biomass fuels based on deep peak load regulation according to claim 8, characterized in that: The regulating unit is an intelligent control valve, wherein the intelligent control valve includes a coal powder bin intelligent control valve and a biomass powder bin intelligent control valve. The coal powder bin intelligent control valve is arranged at the connection between the powder transport pipeline and the coal powder bin, and the biomass powder bin intelligent control valve is arranged at the connection between the biomass powder transport pipeline and the biomass powder bin; In addition, the specific control process of the control unit is: When the combustion state in the boiler has not reached a stable stage, the pulverized coal bin intelligent control valve compares the initial pulverized coal feed amount with the pulverized coal feed amount transmitted by the first acquisition unit after data preprocessing, and performs PID algorithm calculation, and automatically adjusts the valve opening according to the calculation results of the PID algorithm; at the same time, the biomass powder bin intelligent control valve compares the initial biomass powder feed amount with the biomass powder feed amount transmitted by the first acquisition unit, and performs PID algorithm calculation, and automatically adjusts the valve opening according to the calculation results of the PID algorithm, and through the synchronous adjustment of the pulverized coal bin intelligent control valve and the biomass powder bin intelligent control valve, the intelligent adjustment of the pulverized coal and biomass powder feed amounts is realized when the combustion state in the boiler has not reached a stable stage; When the combustion state in the boiler has not reached a stable stage, the pulverized coal bin intelligent control valve compares the optimal pulverized coal feed amount transmitted by the prediction unit with the pulverized coal feed amount transmitted by the first acquisition unit, and performs PID algorithm calculation, and automatically adjusts the valve opening according to the calculation results of the PID algorithm; at the same time, the biomass powder bin intelligent control valve compares the optimal biomass powder feed amount transmitted by the prediction unit with the biomass powder feed amount transmitted by the first acquisition unit, and performs PID algorithm calculation, and automatically adjusts the valve opening according to the calculation results of the PID algorithm, through the synchronous adjustment of the pulverized coal bin intelligent control valve and the biomass powder bin intelligent control valve, thereby realizing intelligent adjustment of the pulverized coal and biomass powder feed amounts when the combustion state in the boiler reaches a stable stage.

10. An intelligent feed adjustment method for a boiler blended with low-carbon biomass fuel based on deep peak shaving, applied to an intelligent feed adjustment system for a boiler blended with low-carbon biomass fuel based on deep peak shaving according to any one of claims 1 to 9, characterized in that: The specific steps are: 1) When the combustion state in the boiler has not reached the stable stage, the specific steps are: S21, the staff sets the initial parameters in the controller according to actual needs and transmits them to the control unit; S22, the first and second acquisition units transmit the data to the data processing unit for preprocessing, and then the first acquisition unit transmits the data to the control unit, and the second acquisition unit transmits the data to the controller; S23, the control unit compares the initial parameters with the data transmitted by the first acquisition unit, and performs PID algorithm calculation, automatically adjusts the valve opening according to the calculation result of the PID algorithm, and then transports the evenly mixed coal powder and biomass powder to the boiler through the transport feeding unit for combustion; at the same time, the controller determines whether the combustion state in the boiler has reached a stable stage based on the data uploaded by the second acquisition unit; S24, repeat steps S22-S23; S25, when the combustion state in the boiler reaches a stable stage, the controller 14 sends instructions to the second and third acquisition units to transmit data to the data processing unit, and stops executing step S24; 2) When the combustion state in the boiler reaches a stable stage, the specific steps are: S31, the first, second and third acquisition units transmit data to the data processing unit for preprocessing, then the first acquisition unit transmits the data to the control unit, and the second and third acquisition units transmit the data to the prediction unit; S32, the prediction unit predicts NO based on the neural network prediction model and optimization algorithm x The optimal coal powder feed amount and the optimal biomass powder feed amount with the minimum concentration are then transmitted to the control unit; S33, the control unit compares the data transmitted by the prediction unit with the data transmitted by the first acquisition unit, and performs PID algorithm calculation, automatically adjusts the valve opening according to the calculation result of the PID algorithm, and then transports the evenly mixed coal powder and biomass powder to the boiler through the transport feeding unit for combustion; S34, repeating steps S31-S33 to complete the intelligent adjustment of the feed amount of pulverized coal and biomass powder when the combustion state in the boiler reaches a stable stage; S35. After the boiler stops burning, stop executing step S34.

Images